Digital audio file formats

Digital audio file formats are used to store digital audio data on a computer system. A codec performs the encoding and decoding of the raw audio data while this encoded data is stored in a container file.

Before we begin this discussion on digital audio file types, we need to clarify a few points before we get started…

  1. Please read my post on hi-res audio to understand what it refers to
  2. Audio waves are analogue. The closer digital audio file formats get to representing that the better.
  3. Lossless compression is a class of data compression that allows the original data to be perfectly (or almost perfectly) reconstructed from the compressed data. They try to NOT “lose” any data and are thus the best digital audio file types
  4. The downside to the very best high-resolution file formats is their size. It has a huge knock-on effect in terms of equipment needed, storage units needed and so on.
  5. Computing devices made by Apple and/or PC manufacturers are rarely optimised purely for sound. 
  6. By contrast, lossy compression permits reconstruction only of an approximation of the original data, though usually with greatly improved compression rates. This involves discarding data and as such, the audio does not sound as good. Lossy compression is an irreversible compression – in other words, it loses data permanently.
  7. The highest quality MP3 has a bitrate of 320kbps. CDs have a quality of about (16-bit, 44.1kHz). True high-resolution files have a much higher sampling rate and range from 24-bit/96kHz or 24-bit/192kHz or above. Typically, the higher the sampling rate the better the quality of the audio captured (as there is more data).

There are many different digital audio file types including ALAC, WAV, FLAC, DSD, MP3 and MQA, etc. Let us start with MQA (Master Quality Authenticated). The title “sounds” very impressive, but is it?

A typical MQA audio file is usually 44.1khz or 48khz in size (CD quality). It will be three times smaller in file size than a FLAC of the same audio which tells you a lot. MQA isn’t lossless. It is lossy and is obviously worse than a FLAC.

MQA appears to be a rather effective scheme for generating licensing fees. Rather curiously, some high-end Hifi manufacturers do not use MQA Decoders in their high-end products, as it won’t give you optimal sound quality and adds an unnecessary cost. I personally stay away from them.

Apple Macbooks/Ipads are not designed to be an audiophile’s dream. Itunes plays Mp3 files. They are convenient for storing large music libraries on Apple devices. But, AAC, Mp3, APE and Ogg Vorbis digital audio file formats use lossy compression. They all have a rather “ropey” sound quality. Here are some typical complaints…

  • They have added distortion
  • They can sound flat and two dimensional – They have lost a lot of the subtle, delicate stuff, like ambience, space and realism
  • They can sound fuzzy, muddled and mushy.
  • They can miss or cut out track sections. For example, “Abbey Road” has several songs stitched together. Lossy files can ruin this effect altogether.

A Hi-Res Digital audio file format has a bitrate (9,216 kbps). It is nearly seven times higher than that of CDs (1,411 kbps) and almost 29 times higher than that of MP3s (320 kbps). And the higher the bitrate, the more accurately the signal is measured. So there is a massive jump in sound quality between true hi-res lossless files and rather ropey lossy files.

Hi-res digital audio file formats

Direct Stream Digital (DSD) is a trademark used by Sony and Philips for their system for digitally encoding audio signals for the Super Audio CD (SACD). It is a niche digital music format for audiophiles. Digital signals take the form of two states that can be represented by 0s and 1s, but this information needs to be arranged in a specific way to be of any use.

In just about every case, the system to organise the digits is known as PCM (Pulse Code Modulation). With PCM, the original analogue music waveform is described in two parts. The first is its amplitude (size). In CD this is represented by 16-bits of digital data, which gives us the ability to define 65,536 different signal levels.

The original music waveform has to be measured at regular intervals in order for it to be represented properly. The measurement is done 44,100 times a second. While that looks like an arbitrarily large number, it’s chosen quite carefully to ensure that the full frequency range of human hearing (20Hz to 20kHz) is covered.

It’s a hi-res format that produces a signal in a different way to that employed by the PCM system that can be transmitted as WAV, FLAC, ALAC or AIFF. If we were to look at a ’24 bit 96kHz’ file- a commonly used high-resolution sample rate, this contains a stream of information 24 bits in size. DSD takes a different approach to the creation of a high-resolution audio signal.

Instead of using many bits of information in the signal, DSD uses a single bit. However, instead of sampling the information several thousand times a second, this single bit samples 2.8 million times a second to generate the audio signal.

FLAC (Free Lossless Audio Codec) is a digital audio file format that offers bit-perfect copies of CDs but at half the size and is royalty-free. The format emerged in 2001 and now, almost every record label is on board with it. The sound quality offered by FLAC comes very close to studio masters, according to the Society of Sound.

Apple Lossless Audio Codec (ALAC) is an alternative to FLAC designed by Apple to play on Apple devices. FLAC and ALAC are about the same in terms of sound quality. I-Tunes does not play FLAC files. But, if you want to use I-Tunes as a hi-res music server, you can do it with ALAC files.

Both WAV and AIFF are uncompressed digital audio file formats, which means they are exact copies of the original source audio. The two formats are essentially the same quality; they just store the data a bit differently. AIFF is made by Apple, so you often see it a bit more on Apple products and WAV is more common on Windows. However, since they’re uncompressed, they take up a lot of unnecessary space.

WMA Lossless (hi-res)is a lossless incarnation of Windows Media Audio; but is no longer well-supported by smartphones or tablets.

APE is, in one way, the most efficient compressed lossless file because it has the most compression without losing audio quality. The drawback is that it has very limited compatibility with standard systems/platforms.

Disadvantages of FLAC

  • FLAC downloads have one major competitor: streaming. Some streaming services use lower quality audio files because they are quicker to download. However, the FLAC will still be available and is unlikely to disappear anytime soon.
  • FLAC is not directly supported by Apple (But you can easily get around this with the right software. I use MplayerX, Elmedia Player and Vox.)
  • FLAC files are compressed when they are encoded and then uncompressed when decoded. This could result in a slight loss of data. Uncompressed digital audio file formats such as WAV, AIFF and DSD will not have this issue.

Advantages of uncompressed lossless audio

If you can get hold of an original DSD file straight from the recording studio; one that took great care with that recording, used extremely good musicians and had a good digital setup (microphones, room, etc), then…

  • It is possible to draw the analogue waveform simply by looking at the density of 0s and 1s in the DSD digital stream. The DSD sound wave is thus likely to be a perfect replica of the analogue recording.
  • DSD files will include some sensational music and have a better-detailed soundstage.
  • You will find that some of the finest performances by orchestras and composers have been captured in DSD (thanks to the care and effort that went into them). They sound absolutely fantastic. DSD is thus a popular choice for well-recorded Jazz and Classical music.

WAV and AIFF

  • WAV files have information about the file including the mono or stereo property, bit depth, sample rate and the number of tracks. Unless you’re editing the audio, you probably don’t need this information.
  • WAV allows us to listen to the highest recording rates with tremendous dynamic ranges (up to 192 kHz). AIFFs are Apple’s equivalents to WAV files.
  • The above points have made them excellent formats to work with studio records, mastering, and advanced audio applications.

Disadvantages of uncompressed lossless audio

  • DSD is different in that its underlying audio storage is pulse-density modulation (PDM). This makes certain demands on your playback equipment for “true” PDM support, although other mechanisms exist to convert the DSD signal to PCM for broader DAC support.
  • There are far fewer bands/publishing labels that actually offer DSD recordings. DSD is NOT a mainstream format. FLAC usage is widespread.
  • Due to its high-quality codec, DSD is impractical for streaming. WAV and AIFF files are also impractical for streaming due to their file sizes and general availability
  • You would probably have to do things in a specialised way to use DSD, WAV or AIFF files, e.g. subscribe to websites that offer these files (expensive due to the storage issues) and have a dedicated ethernet connection, etc. This is not always the case with FLAC.
  • A low-resolution track upscaled to DSD, WAV or AIFF will not sound any better!
  • DSD is, thus, not always magically better than its rivals. A ‘standard’ DSD file – often referred to as DSD64 is roughly equivalent to a sample rate of 24/88.2kHz. ‘Double DSD’ or DSD128 samples that single bit of information 5.6 million times a second to give you a signal equivalent to 24/176.2kHz. Again, this is a sample rate that can be reproduced by formats that are not DSD. Higher rates exist but they are very, very rare and you really do have to hunt them down.
  • It’s not very practical to manipulate a DSD recording. Many things are required post-recording such as equalisation, editing, dynamic range control and adding reverb, etc This usually involves the DSD stream being converted to PCM for processing and then back again. That’s hardly a pure way of doing things.
  • DSD files can have issues with background noise too. FLAC, WAV and AIFF files are often cleaner.
  • WAV, DSD and AIFF files are very large. To compare, a CD recording (44.1 kHz, 16-bit) is ~30 MB on average, while standard WAV can take up to 500 MB. Mari Kodama’s pure DSD-recorded Beethoven Piano Sonatas (16 and 18), a beautiful recording, is 2.7GB and you only get 9 tracks.
  • All digital files can get corrupted and/or be difficult to play. These issues can easily be overcome by changing the software used to play them and/or by downloading a better file. This is again easier to do with FLAC.
  • Metadata is not standardised in WAV therefore some metadata may be lost in the conversion to WAV.
  • Uncompressed audio formats can seriously affect every aspect of your setup. You will need to carefully consider the quality of Hifi components used, the quality of internet provided, the software required and whether or not it deserves a dedicated room with proper acoustics, etc.
  • For DSD, AIFF and WAV files, it makes sense to use specialist websites with reviews, subscribe to them, read thoroughly to find the best recordings and download them over an ethernet cable. This will be both time-consuming and ludicrously expensive.
  • If you then compare the downloaded file with the FLAC of the same recording, you could actually disagree with the original recommendation. But, you will have lost both time and money finding that out.

Which digital audio file formats are the best?

  • All lossless files have excellent sound quality. So much depends on what system you buy, what kind of music you like, how much money you are willing to spend and your personal preferences, etc. Some people will prefer a DSD recording of artist A and then prefer the AIFF for artist B, etc. System configuration can also affect the type of file that sounds best on it and the recording can have a massive impact too, etc. Finding which lossless file works best on your system will be a nightmare. It isn’t practical. Everyone simply gets what they can.
  • FLAC, AIFF and DSD digital audio file formats support the inclusion of metadata. So if you have a Hifi system with a screen for displaying it, you can learn a lot about the recording and/or have an aesthetically pleasing display (Album cover, artist, track title, length of the song, etc.) that encourages you to listen to more music. If you are streaming the music from Spotify, for example, you may also see the Spotify logo on your screen too.
  • If you are very rich, have time to spend finding the files and really want the very best sound, well-chosen DSD, AIFF and WAV files could give you the best sound available – if it really, really matters!
  • FLAC beats uncompressed lossless formats on file size by up to 60-70% at times. That is a truly massive saving and will have a massive knock-on effect in terms of storage space and everything else too.
  • You can easily get hold of FLAC files for free on torrenting websites such as 1377x This is not the case for DSD, WAV and AIFF. You will probably pay through the nose to get them, store them and play them.
  • Why would you waste time and spend, literally piles of money finding files, storing them and buying equipment to play AIFF, DSD or WAV digital audio file formats? FLAC files are also lossless. They may not be quite so perfect. Can you really tell the difference? For most people, FLAC files are good enough…
  • FLAC files simply have very few issues and when they arise, they can usually be resolved easily. This is less likely with DSD.
  • DSD, WAV and AIFF digital audio file formats are not practical for most people. FLAC files are simply far more accessible. What HiFi states they are “considered the preferred format for downloading and storing hi-res albums.” They are the clear winner.
  • I personally use FLAC digital audio file formats for the music I listen to the most and MP3 files for everything else.

If you are serious about hi-res music, you will want to invest in a good quality digital to analogue converter. So, for example, the Cambridge Audio CXN (V2) network streamer has been designed to process a wide array of digital audio file formats. This means you can connect to a dedicated Network Attached Storage (NAS) device such as the Synology Diskstation DS220j NAS.

Synology recommends their “Media Server” package for UPnP (Universal Plug and Play) and others recommend Minimserver software. Alternatively, you can also use Plex which is a good I-tunes alternative. Plex organizes all of your personal media so you can easily access and enjoy it.

Plex supports MP3, M4A, FLAC, and WMA digital audio file formats. The Plex Media Server for Mac/PC seamlessly connects your Plex clients with all of your local and online media. The combination of centralized library management, streaming of online content, and powerful transcoding functionality provides an unrivalled level of flexibility and ease of use.